No abstract

The association of tree species with particular soil types contributes to high b diversity in forests, but the mechanisms producing such distributions are still debated. Soil nitrogen (N) often limits growth and occurs in differentially available chemical forms. In a Bornean forest where tree species composition changes dramatically along a soil gradient varying in supplies of different N-forms, we investigated whether tree species' N-uptake and soil specialization strategies covaried. We analyzed foliar 15 N natural abundance for a total of 216 tree species on clay or sandy loam (the soils at the gradient's extremes) and conducted a 15 N-tracer experiment with nine specialist and generalist species to test whether species displayed flexible or differential uptake of ammonium and nitrate. Despite variation in ammonium and nitrate supplies and nearly 4 % difference in foliar d 15 N between most soil specialists and populations of generalists on these soils, our 15

No abstract

Second generation biofuels, like cellulosic ethanol, have potential as important energy sources that can lower fossil fuel carbon emissions without affecting global food commodity prices. Agricultural crop residues, especially maize, have been proposed for use as biofuel, but the net greenhouse warming effect of the gained fossil fuel carbon offset needs to account for any ecosystem carbon losses caused by the large-scale maize residue removal. Using differential 13 C isotopic ratios between residue and soil in an incubation experiment, we found that removal of residue increased soil organic matter decomposition by an average of 16%, or 540-800 kg carbon ha À1 . Thus, removal of residue for biofuel production can have a hidden carbon cost, reducing potential greenhouse gas benefits. Accurate net carbon accounting of cellulosic biofuel needs to include not only fossil fuel savings from use of the residue, but also any declines in soil carbon caused directly and indirectly by residue removal.

Soil seed banks play an important role in the distribution and composition of plant communities in semiarid grassland ecosystems. However, information on how spatial scale influences the spatial heterogeneity of soil seed banks in a grassland under grazing disturbance is still lacking. Based on field sampling and greenhouse germination, we measured the species composition and seed density of soil seed banks at different spatial scales (30 m×30 m, 30 m×60 m and 30 m×90 m) along a topographical gradient in a sandy grassland in Horqin Sand Land, Northern China. By applying geostatistical methods, we examined how spatial scale and topography affected the spatial distribution of soil seed banks in the study area. Our results showed that the total number of species in soil seed banks, as well as the number of dominant annuals, increased with the increase of spatial scales. Seed density in soil seed banks decreased with the increase of spatial scales due to an increase in the slopes and relative heights of the sampling points. Geostatistical analysis showed that the relative structural variance (C/(C 0 +C)) of seed density and species richness were over 65% for all spatial scales, indicating that these variables had an obvious spatial autocorrelation and the spatial structured variance accounted for the largest proportion of the total sample variance. Spatial autocorrelation of seed density in soil seed banks increased with the increase of measured scales, while that of species richness showed a reverse trend. These results suggest that the total number of species in soil seed banks is spatial scale dependent and lower topography may accommodate more seeds. Spatial distribution of seed density in soil seed banks is also scale dependent due to topographic variation. Grassland management, therefore, needs to consider local grazing disturbance regime, spatial scale and topography.